A5101430203- Remote Sensing and LiDAR Applications
- Robotics and Sensor-Based Localization
- 3D Surveying and Cultural Heritage
- Remote Sensing and Land Use
- Landslides and related hazards
- Robotic Path Planning Algorithms
- Land Use and Ecosystem Services
- Synthetic Aperture Radar (SAR) Applications and Techniques
- Topic Modeling
- Remote Sensing in Agriculture
- Impact of Light on Environment and Health
- Environmental Changes in China
- Text and Document Classification Technologies
- Sentiment Analysis and Opinion Mining
- Optical measurement and interference techniques
- 3D Shape Modeling and Analysis
- Software Engineering Research
- Forest ecology and management
- Natural Language Processing Techniques
- Geotechnical Engineering and Analysis
- Tree Root and Stability Studies
- Advanced Image and Video Retrieval Techniques
- Control and Dynamics of Mobile Robots
- Urban Heat Island Mitigation
- Cryospheric studies and observations
Chang'an University
2019-2025
Chinese University of Hong Kong
2024
Jilin University
2022
State Key Laboratory of Hydraulics and Mountain River Engineering
2018
Nanjing University
2012-2018
Sichuan University
2018
Jiangsu Center for Collaborative Innovation in Geographical Information Resource Development and Application
2018
Terrestrial laser scanning (TLS) and mobile (MLS) data can be used to obtain abundant precise side information of trees. Therefore, it enable extracting individual tree parameters, such as the height, crown size, base diameter at breast provide basic for forest research management. This study proposes a technical framework segmenting trees from TLS MLS data. contains six steps: 1) preprocessing, 2) octree construction, 3) spatial clustering, 4) stem detection, 5) initial segmentation, 6)...
The Heifangtai Loess terrace in northwest China is frequently affected by landslides due to hydrological factors, establishing it as a significant research area for loess landslides. Advanced time-series InSAR technology facilitates the retrieval of surface deformation information, thereby aiding monitoring landslide status. However, existing methods that analyze patterns do not fully exploit displacement time series derived from InSAR, which hampers exploration potentially coexisting within...
A complete picture of the deformation characteristics (distribution and evolution) geotechnical infrastructures serves as superior information for understanding their potential instability mechanism. How to monitor more completely accurately these (either artificial or natural) in field expediently roundly remains a scientific topic. The conventional monitoring methods are mostly carried out at limited number discrete points cannot acquire data whole structure. In this paper, new methodology...
Change detection is a major issue for urban area monitoring. In this paper, new three-step point-based method detecting changes to buildings and trees using airborne light ranging (LiDAR) data proposed. First, the LiDAR from two dates are accurately registered iterative closest point algorithm, progressive triangulated irregular network densification filtering algorithm used separate ground points non-ground points. Second, an octree generated store index irregularly-distributed Finally, by...
As a fundamental capability of mobile robots, path planning highly relies on the accurate localization robot. However, limited consideration for localizability (which describes acquiring localization) has been made in planning. This brings high risk choosing that is optimal but results robot easily getting lost. There exist two key challenges to address this problem: 1) How evaluate and its impact 2) balance standard criteria (e.g., shortest travel distance, obstacle-free path, etc.. To...
Interactive argument pair identification is an emerging research task for mining, aiming to identify whether two arguments are interactively related. It pointed out that the context of essential improve performance. However, current context-based methods achieve limited improvements since entire typically contains much irrelevant information. In this paper, we propose a simple contrastive learning framework solve problem by extracting valuable information from context. This can construct...
Cross-domain text classification aims to transfer models from label-rich source domains label-poor target domains, giving it a wide range of practical applications. Many approaches promote cross-domain generalization by capturing domain-invariant features. However, these methods rely on unlabeled samples provided the which renders model ineffective when domain is agnostic. Furthermore, are easily disturbed shortcut learning in domain, also hinders improvement ability. To solve aforementioned...
<title>Abstract</title> Landslide disasters—prone to river blocking—are widely distributed in the mountainous areas of southwest China and are characterized by high-elevation long-runout movement with great destructive power. Identification monitoring landslides prediction unstable landslide movements significance for regional disaster mitigation prevention. In this study, we used interferometric synthetic aperture radar (InSAR) identify monitor potential Gongjue County, selected Sela as a...
This paper extracted green space distribution information from multi-temporal LANDSAT multi-spectral remote sensing images to analyze the evolution of Nanjing’s between 1988 and 2007. The findings suggest that current is not balanced total area has decreased over time. trend towards fragmentation obvious in disproportionately large decline medium-sized spaces. Mountain forest other spaces, however, have been better protected. Before 1998, was most apparent spaces city’s inner circle. After...